The serological diagnosis of virus infections depends for the most part on neutralization and complement fixation tests. Both of these have been described in the case of herpes simplex infections, but the neutralization test is used almost exclusively at present, owing to the conflicting results obtained by complement fixation. The reports in the literature on this subject are at variance with each other. Kraus andTakaki (1925, 1926) and Takaki, Bonis, and Koref (1926) reported positive fixation with herpes virus using convalescent human and rabbit sera with a rabbit brain antigen. These workers also reported that they were able to distinguish by this method between herpes and other neurotropic viruses such as vaccinia, rabies, and the virus of Japanese B encephalitis. Their work, however, was not confirmed by Greenbaum and Harkins (1925), Tang and Castaneda (1929), and Gay and Holden (1929). Schultz and his colleagues in a series of investigations (Schultz, 1928; Schultz, Bullock, and Lawrence, 1928; Schultz and Hoyt, 1928) not only failed to demonstrate complement fixation with herpes but stated that the in vitro combination of antigen and antibody in the case of filterable viruses did not occur, and that the degree of fixation reported by previous workers was due, first to bacterial contamination of the virus antigens, and secondly to a lack of specificity due to inadequate controls. Bedson and Bland (1929) were able to demonstrate specific complement fixation with herpes by using an immune guinea-pig serum and an antigen prepared from guinea-pig pads. This method eliminated the nonspecific and anticomplementary results produced by rabbit brain antigen and rabbit serum. Brain (1932) using the same method as Bedson and Bland was able to show that there was a definite relationship between the presence of complementfixing antibodies and neutralizing antibodies in human sera. Myers and Chapman (1937) found no satisfactory evidence of complement fixation between several different herpes antigens and immune herpes sera, in marked contrast to the results they obtained with vaccinia and virus III. Their results were either completely negative, non-specific, or showed fixation only at a low titre. The discrepancy in these findings was probably due to several factors; first, the lack of a sufficiently